Clinical neuroscience studies have suggested that dysfunctional reactivity of the brain circuits mediating emotion may be a key factor in the etiology and maintenance of many psychiatric disorders. It is also well established that the sensory cortical response to emotional stimuli is an important part of a cascade of events - physiological, cognitive, and behavioral - that define emotional reactivity in humans. For instance, anxiety patients show visual responses that are both biased towards threat cues and lacking discriminative accuracy. Mechanistic knowledge is needed that addresses the question of how such perceptual biases towards threat features are acquired (and unlearned) in the human visual system. This has been difficult because reliable methods to quantify single trials of neural activity are not available at this time. In this multidisciplinary research project, we propose to use novel computational and experimental approaches to fill this gap. We aim to objectively characterize and quantify - on a trial by trial basis - the temporal evolution of neura changes in the human visual system that accompany the acquisition and extinction of conditioned fear. An objective and reliable description of the time course of visual changes during fear learning will assist in ongoing efforts aiming to develop objective diagnostic categories of fear disorders, to understand and quantify effects of treatment, and to develop new forms of attention/perception trainings in the fear and anxiety disorders. PUBLIC HEALTH RELEVANCE: Many psychiatric disorders are characterized by heightened attention and perception of threat-related visual information, and such perceptual biases are also seen in healthy observers facing a threatening visual stimulus. Using novel methods in biomedical engineering and neurophysiology, this multidisciplinary research project examines how healthy observers acquire perceptual biases during classical fear conditioning, and how these biases can be eliminated through extinction learning. Finding reliable and valid quantitative neural parameters of fear learning is important for the objective assessment of brain circuitry underlying anxiety disorders in humans, and it also has the potential of establishing neurofeedback treatments whose aim is to alter the functional neuroanatomy of fear.